PROJECT SUMMARY Triple negative breast cancer (TNBC) is a major subtype of breast cancer that is associated with generally poor outcome, younger age at diagnosis, and worse prognosis than other subtypes. It currently has no options for targeted therapy, and chemotherapy remains the only pharmacologic option; thus, better treatments are urgently needed. We recently demonstrated the potential of a class of epigenetic targeted agents, BET inhibitors, as a promising new therapy in TNBC. However, the rapid development of resistance necessitates further study to determine how to effectively administer these drugs and to prevent resistance. Treatment with the prototypical BET inhibitor, JQ1, is known to induce phenotypic changes in cellular state, which may mediate resistance. The high degree of intratumor heterogeneity in TNBC may also contribute to development of resistance and treatment failure. Therefore, a better understanding is needed of how tumor populations are affected by the selective pressures of treatment. Furthermore, effective combination therapies must be developed and their administrations optimized in order achieve a more durable response. I hypothesize that double and triple combination therapies with paclitaxel chemotherapy and PD-L1 immunotherapy, as well as the order in which combinations are administered, will have an effect on treatment efficacy and tumor evolution. In Aim 1, I will investigate population dynamics in response to JQ1 combination therapy with paclitaxel, using TNBC cell lines barcoded with a high-complexity DNA library to track individual subclones during treatment, both in culture and in xenografts in immunodeficient mice. In Aim 2, I will investigate optimal administration of JQ1 with paclitaxel by testing concomitant and sequential therapy in both orders in xenografts with barcoded cells, in order to assess differences in population dynamics between treatment schedules and differences in cellular response with single cell RNA-seq. I will then use EvoSeq, a method to isolate cells from within a population with barcode-level specificity, to retrieve cells with differentially selected barcodes between treatment schedules and examine whether they are predisposed to resistance and whether pretreatment with one drug alters sensitivity to the second drug. In Aim 3, I will characterize changes in heterogeneity that result from the triple combination of JQ1, paclitaxel, and anti-PD-L1 antibody, after testing all single agents, double combinations, and triple combination in MMTV-PyMT cells in immunocompetent mice. The results of this study will lead to a deeper understanding of how intratumor heterogeneity modifies treatment response and establish how to optimally combine BET inhibitors. This research will also have direct translational impact in informing the rational design of clinical trials and, if successful, would lead to improved survival for patients with TNBC.